Back to content

Guide • Scouting playbook

The Orchard Scouting Playbook

A full guide on how to sequence satellite, drone, and ground-truth scouting for lower-cost field decisions and better yield protection.

A Multi-Layer Remote Sensing Framework for Vineyard and Berry Scouting in British Columbia

1. The Myth of the Magic Map

On the limitations of single-dashboard assumptions in field scouting

Single-dashboard monitoring claims remain common in agricultural technology messaging, but operational outcomes in heterogeneous fields are mixed.

Field conditions in specialty crops rarely support one-layer inference.

In 2024, British Columbia's fruit industry learned this lesson the hard way. When a severe January cold snap plunged temperatures to -30°C in Kelowna, it wiped out 84% of wine grape production and devastated stone fruit orchards across the Okanagan and Similkameen Valleys [Agriculture and Agri-Food Canada, 2025]. No single dashboard could have predicted which vines would survive and which blocks needed immediate replanting. The growers who recovered fastest used multiple data sources—not magic maps.

The Reality of Multi-Layer Scouting

In vineyards and berry farms, satellite, drone, and phone images do different jobs. None replaces the others. The fastest way to waste time is to use the wrong tool for the wrong question.

The short answer:

Satellite (Sentinel-2, 10m resolution) is good for broad change detection and deciding where to look first. According to the Copernicus programme, Sentinel-2 provides 12 spectral bands at 10-20m spatial resolution with a 5-day revisit frequency [Copernicus, 2024]. A single 10m pixel covers 100 square meters—enough to blend multiple vineyard rows, alleyways, and soil backgrounds into one averaged signal.

Drone is good for block-level detail and showing the shape of a problem. Research from Penn State Extension confirms that drone imagery captures "a few centimeters square per pixel"—dramatically higher resolution than satellite [Penn State Extension, 2024]. This is the difference between knowing "something is off in Block 3" and seeing exactly which rows show water pooling.

Phone or ground photos are what you need when it's time to confirm what the issue actually is. Only ground verification can distinguish between winter injury, disease pressure, and irrigation problems that all look similar from above.

The Cost of Getting It Wrong

If Sentinel-2 outputs are interpreted as diagnosis rather than triage, decision error probability increases. A 2024 study on crop type mapping with Sentinel-2 at 10m scale highlighted the fundamental challenge: pixel values differ based on polygon sizes and mixed land cover, making single-pixel interpretation risky for row crops [Khan et al., ResearchGate 2024].

If drone deployment is used as a default for routine checks, time and cost escalate rapidly. Entry-level agricultural drones cost $2,000-$5,000, while advanced spraying drones exceed $15,000-$20,000 [UAV Coach, 2025]. The agricultural drone market reached $3.3-6.4 billion globally in 2024, driven by demand for targeted—not blanket—applications [Market Research Future / IMARC Group, 2024].

Without ground verification, causal inference remains uncertain, and intervention risk increases.

The Stack Decision Framework

Where should I look?

What's the shape/extent?

What is this really?

Start with the question

What do you need to know?

Satellite
10m resolution
5-day revisit

Drone
cm-level resolution
Block-level detail

Phone/Ground
Visual confirmation
Diagnosis

Flag unusual zones

Map problem extent

Confirm cause

Prioritize blocks

Plan intervention

Execute treatment

Use satellite when you want: Cheap, repeatable field-wide signal over time Monitoring across many sites simultaneously Broad spatial variability detection Change detection after weather events

Use drone when you need: Detail inside a specific block Centimeter-level precision for row crops Documentation for consultants or insurance Mapping of variability at actionable scale

Use phone when you need: Proof and context for diagnosis Visual confirmation before expensive treatments Evidence of specific symptoms (leaf color, cane damage, fruit condition)

The Bottom Line

The "one dashboard" myth persists because it's easier to sell than complexity. But BC's 2024 cold snap proved what experienced growers already knew: different questions require different tools.

Over the next four installments, we'll break down exactly where each tool shines, where it breaks, and how to stop making expensive mistakes with your imagery.


Sources (all 2024-2025): Agriculture and Agri-Food Canada. (2025). Statistical overview of the Canadian fruit industry, 2024. Government of Canada. Copernicus. (2024). Sentinel-2 for Agriculture. European Space Agency. Khan, A. et al. (2024). Mapping Crop Types At a 10 m Scale Using Sentinel-2 Data and Machine Learning Methods. ResearchGate. Penn State Extension. (2024). Unmanned Aerial Vehicle-Based Crop Scouting in Fruit Trees. UAV Coach. (2025). Drones in Agriculture: The Best Agricultural Drones of 2025. Market Research Future / IMARC Group. (2024). Agriculture Drones Market Size, Share & Forecast.


2. Satellite Scouting: The Art of Triage

Signal versus certainty: What Sentinel-2 can actually tell you

Satellite imagery is free, global, and available every 5 days. It is also limited to 10-meter pixels that blend everything in their path. Understanding this trade-off is the difference between useful monitoring and expensive confusion.

When Satellite Works

Satellite is useful when the question is simple:

Which fields or blocks changed? Where should I send someone first? Is this area consistently weaker than the rest over time? Did this stress pattern show up after heat, rain, smoke, irrigation issues, or a management event?

For BC vineyards and berries, satellite is strongest for: Routine monitoring across many sites: With 12,555 hectares of grape vineyards and 11,911 hectares of highbush blueberries in BC (2024 data), no grower can physically scout every block weekly [Statistics Canada, 2025]. Broad spatial variability detection: Sentinel-2's 12 spectral bands can identify zones with consistently different vegetation signatures [Copernicus, 2024]. Comparing current conditions to recent history: The 5-day revisit frequency enables trend analysis that single snapshots cannot provide. Flagging zones that deserve closer inspection: Satellite identifies priority zones; it does not provide final causal diagnosis.

The Hard Limit: 10 Meters is Coarse for Row Crops

This is the part people love to hand-wave away.

A 10m pixel covers 100 square meters. To understand what this means in practice:

In vineyards: One Sentinel-2 pixel can easily blend: Multiple vine rows (typically 2-3m spacing) Alleyways and drive rows Missing vines or replant gaps Shadows from trellis systems Soil background variability Edge effects from adjacent blocks

In berries: That same pixel can mix: Canopy from healthy plants Wheel tracks and bare ground Irrigation variability Weed patches Stand gaps from winter damage

A 2024 arXiv study on hierarchical crop classification noted that even with machine learning, Sentinel-2 at 10m resolution struggles with fine-grained discrimination in heterogeneous agricultural landscapes [arXiv, 2025]. The study integrated 30m hyperspectral data with 10m Sentinel-2 time series to improve classification—acknowledging the fundamental resolution limitation.

Use It for Signal, Not Certainty

What satellite can reliably detect: Large-scale patterns (multiple hectares) Persistent anomalies over multiple revisits Seasonal trends in vegetation vigor Changes following major weather events Block-to-block differences in overall performance

What satellite misses: Early, plant-level symptoms Narrow row-specific issues Under-canopy or hidden conditions Small problem patches (sub-pixel scale) Issues masked by mixed pixels Anything obscured by clouds, haze, smoke, or timing gaps

Canada's 2024 fruit production data illustrates this limitation. While Quebec achieved record cranberry and apple yields, BC's grape production fell 84% due to the January cold snap [Statistics Canada, 2025]. Satellite imagery in February might have shown reduced NDVI in affected blocks—but it could not distinguish between dead vines, dormant healthy vines, and vines with delayed bud break. Only ground inspection revealed the true damage.

Cost and Speed Analysis

Factor Satellite (Sentinel-2) Assessment
Cost per revisit Free (Copernicus programme) Excellent
Coverage speed Global, 5-day revisit Excellent
Spatial resolution 10m Poor for row crops
Temporal resolution 5 days (cloud permitting) Good
Best use case Monitoring and triage Excellent
Diagnosis capability Limited Poor

The Workflow: Satellite as Triage Tool

No

Yes

Low

High

Critical

Weekly/Monthly
Satellite Pass

NDVI/NDRE
Anomaly?

Continue Monitoring

Flag for
Priority Scouting

Severity?

Add to next
routine ground check

Deploy drone
same week

Immediate
ground visit

Real-World BC Context

In 2024, British Columbia's highbush blueberry production rebounded 41% year-over-year despite challenging weather [Agriculture and Agri-Food Canada, 2025]. Growers using satellite monitoring could track: Bloom timing differences across varieties (visible as NDVI shifts) Areas where late August rains affected fruit quality Blocks with consistently lower vigor throughout the season

What they could not see from satellite: Scorch virus symptoms (visible on individual leaves) Spotted Wing Drosophila pressure (requires trap monitoring) Fruit firmness and sugar content Individual cane health

Key Takeaways

  1. Satellite identifies priority zones; it does not provide final causal diagnosis. Use it to prioritize where you look, not to diagnose what you find.

  2. 10m resolution is coarse for row crops. A single pixel averages too much to reveal plant-level problems.

  3. Temporal consistency beats single snapshots. The value is in tracking change over time, not interpreting one image.

  4. Clouds are your enemy. The 5-day revisit is theoretical; actual usable imagery depends on weather.


Sources (all 2024-2025): Statistics Canada. (2025). Fruit and vegetable production, 2024. Agriculture and Agri-Food Canada. (2025). Statistical overview of the Canadian fruit industry, 2024. Copernicus. (2024). Sentinel-2 for Agriculture. European Space Agency. arXiv. (2025). Fine-grained Hierarchical Crop Type Classification from Integrated Hyperspectral EnMAP Data and Multispectral Sentinel-2 Time Series.


3. Drone Scouting: Detail Without Diagnosis

When to fly, what you'll see, and why pretty maps aren't prescriptions

Drone imagery bridges the gap between satellite's broad brush and ground truth's narrow focus. But centimeter-level resolution comes with its own costs—and its own blind spots.

When Drone Makes Sense

The question changes from "where?" to "what exactly?" and "how big?":

Where exactly is the problem inside this block? How extensive is the affected area? Is the pattern tied to rows, irrigation lines, drainage, slope, or management zones? Do I need imagery detailed enough to act on?

According to Kansas State University's Agronomy eUpdate (April 2025), "drone scouting can be less time-consuming, have more thorough field coverage, and be less labor intensive" compared to traditional ground scouting—when used appropriately [Kansas State University, 2025].

The Resolution Game-Changer

Penn State Extension confirms that drone imagery captures "a few centimeters square per pixel" compared to satellite's meters [Penn State Extension, 2024]. At this resolution, you can often see:

In vineyards: Row-to-row vigor differences Missing or dead vines Canopy gaps and uneven growth Water pooling in low spots Machinery damage to trellis systems Edge effects from adjacent blocks

In berries: Stand variability and weak patches Wheel track compaction patterns Irrigation emitter coverage gaps Drainage problems after rain Weed encroachment boundaries Harvest equipment damage

The agricultural drone market reached $3.3-6.4 billion globally in 2024, with projections showing growth to $12-70 billion by 2035 [Market Research Future / IMARC Group, 2024]. This investment reflects real value—but only when the tool fits the problem.

When to Fly: A Decision Matrix

Location/Extent

What is it?

Is it spreading?

Yes

No

Satellite or Scout
Flags an Issue

Question Type

Fly Drone

Skip Drone
Go to Ground

Pre-Harvest
Inspection

Damage Assessment
Post-Weather

Block-Level
Vigor Mapping

Consultant/
Insurance Documentation

Generate
Actionable Map

Need diagnosis?

Ground truth
required

Map sufficient
for action

What Drone Misses

Despite the impressive resolution, drone imagery has significant limitations:

It is not free. Entry-level mapping drones cost $2,000-$5,000. Professional multispectral units run $10,000-$20,000. Advanced spraying drones like the DJI Agras T50 exceed $15,000-$20,000 [UAV Coach, 2025]. Per-flight costs include operator time, battery cycles, and data processing.

It is not frictionless. Someone must: Plan the flight (weather, wind, airspace) Execute the mission Process the imagery (stitching, georeferencing) Interpret the results Store the data

One flight is a snapshot, not a season. Unlike satellite's 5-day revisit, drone flights are typically event-driven. Comparing "before" and "after" requires deliberate planning.

Patterns without causes. A drone map showing weak vigor in the southeast corner of Block 7 does not tell you why. Is it: Irrigation pressure drop? Soil compaction? Root disease? Winter injury (relevant after BC's 2024 cold snap)? Drainage issue?

Only ground inspection can answer.

Canopy-hidden issues. Drone imagery captures the top of the canopy. It misses: Under-canopy disease pressure Fruit quality issues Cane or trunk damage Root zone problems Internal plant stress not visible from above

Cost and Speed Analysis

Factor Drone Imagery Assessment
Equipment cost $2,000-$20,000+ Significant
Per-flight cost Labor + processing Moderate
Coverage speed 1-2 blocks per flight Slow vs. satellite
Spatial resolution 1-5 cm/pixel Excellent
Temporal coverage Event-driven Limited
Best use case Targeted detail, mapping Excellent
Diagnosis capability None without ground truth Poor

Regulatory Reality

Commercial drone operations in Canada require: Transport Canada Pilot Certificate - Advanced Operations Registered aircraft Compliance with airspace restrictions Insurance coverage

For BC growers near airports (Kelowna, Abbotsford, Vancouver), airspace coordination adds complexity. The 2024 crop disaster response required many emergency assessments—proving that regulatory preparation matters before crisis hits.

Real-World Application: BC 2024 Recovery

Following the January 2024 cold snap that destroyed 84% of BC wine grape production, drone mapping served critical functions:

What drone revealed: Which blocks had surviving vines vs. total loss Spatial patterns of damage (low areas vs. slopes) Priority zones for replanting Documentation for insurance claims

What drone could not reveal: Which vines would recover with pruning Root health status Secondary disease pressure in damaged canopies

Ground crews still had to walk every row to make final replanting decisions. Drone reduced the search area; it did not eliminate the search.

Integration with Satellite Data

The most effective workflow combines both:

  1. Satellite monitoring flags blocks with anomalous NDVI
  2. Drone flights target only those blocks for detailed mapping
  3. Ground inspection validates drone findings before action

This triage approach focuses expensive drone time where it matters most.


Sources (all 2024-2025): Kansas State University. (2025). Agronomy eUpdate, April 3, 2025: Using Drones for Early-Season Field Scouting. Penn State Extension. (2024). Unmanned Aerial Vehicle-Based Crop Scouting in Fruit Trees. UAV Coach. (2025). Drones in Agriculture: The Best Agricultural Drones of 2025. Market Research Future / IMARC Group. (2024). Agriculture Drones Market Size, Share & Forecast. Agriculture and Agri-Food Canada. (2025). Statistical overview of the Canadian fruit industry, 2024.


4. Ground Truth and Expensive Mistakes

Why your phone is the only honest tool, and six blunders that cost you

Phone images are not the cheap substitute. They are the truth check. In an era of $20,000 drones and free satellite data, the most valuable agricultural sensor is still the one in your pocket—used properly.

When Phone/Ground Photos Are Essential

Use ground documentation when you need to answer:

What is this, actually? Is it disease, nutrient deficiency, water stress, herbicide injury, pests, physical damage, or just shadow and soil background? Where exactly is the problem? Is the issue on leaves, fruit, cane growth, crown area, roots, or only on one side of the canopy? Should I act? Before spraying, irrigating, sampling tissue, or bringing in an agronomist—verify.

Ground verification is mandatory when: You need to distinguish one stress cause from another You need leaf, fruit, cane, or crown-level evidence The issue is small, early-stage, or patchy You are deciding whether to invest in intervention You need documentation a grower or insurer can trust

A map tells you where to look. A phone image often tells you what you are looking at. Skip that step and expensive mistakes follow.

The Limitations of Ground-Only Scouting

Ground truth has its own blind spots:

It does not show the whole field. A scout walking representative transects may miss problems in unsampled areas. The 2024 Statistics Canada fruit production report noted that BC's apple crop declined 7% partly due to "significant hail damage in mid-September" that was localized—not uniform across orchards [Statistics Canada, 2025].

It depends on where the scout walked. Human bias toward obvious symptoms means early-stage problems get missed.

It can miss broader spatial patterns. Correctly identifying symptoms in one spot while missing that the pattern correlates with elevation, drainage, or irrigation zones.

It is poor for estimating affected area. "Looks bad in the southwest corner" lacks the quantifiable extent that drone mapping provides.

The Integration Imperative

Ground truth without aerial context can mislead. You may correctly identify downy mildew on leaves in one spot while missing that the pattern follows irrigation pressure drops across the entire field.

The optimal workflow is sequential:

Yes

No

Uncertain

Satellite
Wide-area
Triage

Drone
Block-level
Mapping

Ground
Verification

Diagnosis
Clear?

Execute
Treatment

Expand
Ground Check

Lab Analysis/
Consultant

Six Common Mistakes That Waste Time and Money

1. "Red means disease"

No. A stress map—whether from satellite NDVI or drone multispectral imagery—does not label cause. Low vigor might indicate: Irrigation problems Compaction Shallow soils Nutrient deficiency Trunk damage Drainage issues Disease Replant gaps Shadow effects Mixed soil background

Same red patch. Completely different actions. The only way to know is ground verification.

2. "A clean map means the field is fine"

Also no. A field can look broadly uniform from above and still have: Localized pest pressure (below pixel resolution) Early disease (not yet visible at canopy scale) Fruit quality issues (internal, not spectral) Crown or root problems Irrigation emitter failures (too small for detection)

BC's 2024 grape disaster proved this: many blocks showed normal NDVI through winter dormancy. The -30°C cold snap damage was invisible until bud break—when it was catastrophic.

3. "Higher NDVI means healthier"

Sometimes. Not always. High vigor is not automatically good:

In vineyards: Excessive vegetative growth means: Shading of fruit zones Uneven ripening risk Increased disease pressure from dense canopy Higher labor costs for leaf removal

In berries: Strong canopy signal can hide: Fruit quality issues Disease pressure under dense foliage Over-fertilization wasting inputs

The goal is balanced vigor, not maximum NDVI.

4. "One image tells the story"

Nope. Single-date interpretation is fragile. Context required: Compared to what baseline? When in the season? After what weather? Before or after irrigation? Is the pattern consistent over multiple revisits?

A 2024 study on Sentinel-2 crop classification emphasized the importance of time series over single snapshots for accurate identification [Khan et al., ResearchGate 2024].

5. "Drone replaces ground scouting"

Fantasy. Drone reduces blind wandering. It does not eliminate boots-on-ground verification. Every drone map showing an anomaly should trigger a ground check—until the pattern is understood well enough to predict ground findings.

6. "Satellite is too coarse to be useful"

Wrong. Sentinel-2 at 10m is limited, but still valuable for: Broad variability detection Change monitoring over time Triage across large operations Seasonal trend analysis

The mistake is not using it. The mistake is overclaiming what it can tell you.

Cost of Mistakes: Real Numbers

Canadian fruit farm gate value reached $1.34 billion in 2024 [Statistics Canada, 2025]. BC's share—$338.7 million—represented a 25% decline from 2023 due to the cold snap [Agriculture and Agri-Food Canada, 2025].

In this context, misdiagnosis costs real money: Unnecessary fungicide application: $50-150/acre Delayed replanting decision: lost production for 2-3 years Over-irrigation based on false stress signals: water waste + disease risk Under-fertilization of genuinely weak blocks: yield loss

Ground verification is cheap insurance against expensive mistakes.

The Phone as Documentation Tool

Modern smartphones capture metadata-rich images: GPS coordinates (±3-5m accuracy) Timestamp Camera settings Often, compass orientation

Organized ground photo collections become: Seasonal reference libraries Evidence for insurance claims Training data for AI models Proof of pre-condition for liability disputes

Best Practices for Ground Verification

  1. Sample systematically. Don't just photograph obvious problems. Capture representative healthy areas for comparison.

  2. Document context. Wide shot showing the plant in its row, then close-up of the symptom.

  3. Note metadata. Variety, block, row number, date, recent weather, recent management.

  4. Link to aerial data. Tag ground photos with corresponding satellite or drone imagery dates.

  5. Review seasonally. Compare ground findings to aerial patterns over time to improve interpretation skills.


Sources (all 2024-2025): Statistics Canada. (2025). Fruit and vegetable production, 2024. Agriculture and Agri-Food Canada. (2025). Statistical overview of the Canadian fruit industry, 2024. Khan, A. et al. (2024). Mapping Crop Types At a 10 m Scale Using Sentinel-2 Data and Machine Learning Methods. ResearchGate.


5. The Playbook: A Decision Framework for Vineyards and Berries

The practical workflow that beats any silver bullet

After four installments examining what each tool can and cannot do, here is the integrated workflow that actually works in BC vineyards and berry fields. No magic. No silver bullets. Just a decision framework based on the scale of your question and the stakes of getting it wrong.

Start With the Question

Your Question Start Here
"Where should I look?" Satellite
"What is the shape and extent?" Drone
"What is this issue, really?" Phone/Ground

Then check the stakes:

Situation Approach
Low-cost routine monitoring across many fields Satellite first
Known problem block that needs detail Drone first, then ground verify
Any decision involving intervention Ground verify before acting

The Complete Decision Workflow

Where are the
problem areas?

How big is it?
What's the pattern?

What is it?
Should I treat?

Yes

No

Yes

No

Yes

No

Management
Question

What do you
need to know?

Satellite
Sentinel-2
10m, 5-day

Drone
cm-resolution
Block-level

Ground/Phone
Visual diagnosis
Confirmation

Monitor NDVI/NDRE
across all blocks

Flag anomalies
for priority review

Track trends
over season

Anomaly
significant?

Continue monitoring

Map extent
at row level

Document patterns
for records

Create action
zones

Diagnosis
clear from map?

Execute
treatment

Verify symptoms
in field

Rule out
false positives

Confirm cause
before action

Confident
diagnosis?

Treat/Intervene

Consultant/
Lab analysis

Review outcomes
improve models

Treatment based
on expert ID

Vineyard-Specific Tactics

BC's 12,555 hectares of grape vineyards face unique monitoring challenges—especially after the 2024 cold snap that destroyed 84% of wine grape production in some regions [Statistics Canada, 2025].

Satellite helps with:

Comparing larger blocks over time: Track post-damage recovery across multiple vineyards Finding broad weak zones: Identify which blocks consistently underperform Tracking recurring variability: Spot patterns that repeat season after season Prioritizing blocks for closer review: Focus limited labor where it matters

Example: After the January 2024 freeze, satellite NDVI in April 2024 could identify which blocks had recovered vegetation signals vs. those remaining anomalously low—prioritizing ground inspections.

Drone helps with:

Mapping row-level variability: See individual vine performance Identifying patchy weak canopy zones: Distinguish between uniform damage and scattered losses Seeing drainage patterns: Low areas where cold air settled or water pools Missing vines and replant gaps: Count and locate gaps for replanting decisions Documentation for replanting programs: BC provided $26 million for grape replant programs post-2024 [Agriculture and Agri-Food Canada, 2025]

Phone/ground checks help with:

Confirming bud viability: Satellite and drone cannot distinguish dead buds from dormant healthy ones Checking cane damage: Split bark, internal browning Assessing trunk injury: Cold damage often manifests below ground or in trunk tissue Distinguishing biotic vs. abiotic stress: Disease symptoms vs. winter injury vs. nutritional issues Validating weak map zones: Is low NDVI dead vines, delayed bud break, or just bare soil/alleys?

Vineyard Cautions:

Row crops break simplistic satellite interpretation. The 10m pixel blending row and alley is not a bug—it's a fundamental limitation. Interpreting vineyard NDVI requires understanding that: Trellis architecture creates shadows Alley management (bare soil vs. cover crop) affects signal Row orientation affects sun angle reflectance Missing vines reduce pixel-average NDVI even if remaining vines are healthy

High vigor is not always desirable in wine grapes. Excessive vegetative growth: Shades fruit zones (reducing quality) Increases disease pressure Requires expensive leaf removal Delays ripening

Target: Balanced vigor aligned with production goals, not maximum NDVI.

Berry-Specific Tactics

BC's berry sector—71,497 metric tons of highbush blueberries, 45,988 metric tons of cranberries in 2024—fared better than grapes during the cold snap, with blueberries actually up 41% year-over-year [Agriculture and Agri-Food Canada, 2025]. But berries have their own monitoring challenges.

Satellite helps with:

Triage across many fields: Fraser Valley highbush blueberry growers often manage multiple sites Broad canopy change detection: Spot unusual patterns after weather or management events Monitoring larger planting areas: Cranberry bogs cover significant acreage Finding zones for priority attention: Which blocks deviate from expected seasonal progression?

Drone helps with:

Seeing stand variability: Individual bush performance in blueberries Mapping waterlogging: Critical for cranberries and blueberries in low-lying areas Showing exactly where issues concentrate: Patchy vs. uniform problems Documenting conditions before harvest: Quality assurance records Damage assessment: Post-harvest or weather event evaluation

Phone/ground checks help with:

Confirming disease: Scorch virus symptoms (widespread in 2024), fungal pressure Checking fruit condition: Firmness, color, sugar content—no spectral proxy Cane health assessment: Winter damage in raspberries, blackberries Verifying under-canopy issues: What aerial imagery cannot see Pest monitoring: Spotted Wing Drosophila levels were lower in 2024 but still required trap monitoring [Agriculture and Agri-Food Canada, 2025]

Berry Cautions:

Small patches disappear in coarse imagery. A 10m Sentinel-2 pixel covers approximately 100 blueberry bushes. Early-stage problems affecting 10-20 bushes are invisible from space.

Harvest and traffic damage mimics agronomic stress. Worn paths, equipment compaction, and picking damage can create low-NDVI zones that look like plant stress from above but require different management.

Field edges distort interpretation. Tunnels, hedgerows, and uneven canopy closure create spectral artifacts that can be mistaken for agronomic issues.

The Workflow That Actually Works

For most operations, the useful sequence is not satellite or drone or phone.

It is:

  1. Satellite to monitor and flag unusual zones
  2. Drone when higher-resolution mapping is worth the effort
  3. Phone or ground photos to confirm the cause before acting
Week 1:        Satellite pass → NDVI analysis → Flag Block 7, Block 12
Week 2:        Drone flights → Detailed maps → Southwest corner of Block 7 shows pattern
Week 3:        Ground check → Confirm drainage issue → Install drainage tiles
Week 6:        Satellite pass → Verify improvement → Continue monitoring

That stack is less glamorous than the "one dashboard solves everything" pitch.

It is also a hell of a lot more truthful.

And in agriculture, truthful beats flashy every time.

Integration Checklist

Before investing in any monitoring technology, ask:

[ ] Does this tool answer a specific question I have? [ ] Do I have the workflow to act on what I learn? [ ] Am I verifying aerial findings with ground truth? [ ] Do I understand what this tool cannot tell me? [ ] Is the cost justified by the decisions it enables?

The BC Context: Lessons from 2024

The January 2024 cold snap was a brutal test of agricultural resilience. British Columbia's fruit industry lost an estimated $200+ million in production value [derived from Statistics Canada farm gate value data]. The recovery involved:

$26 million in provincial grape replant programs $92.6 million in Production Insurance and AgriStability payments 306 licensed wineries facing constrained 2024 vintage production Multi-year recovery timelines for perennial crops

In this context, the difference between a tool that finds problems and a workflow that solves them is not academic—it is economic survival.

Satellite monitoring helped prioritize which blocks to assess first. Drone mapping documented damage extent for insurance and replanting programs. Ground verification ensured replanting decisions were based on actual vine mortality, not spectral proxies.

No magic map predicted the cold snap. But the growers with robust, multi-layer monitoring systems recovered faster—because they knew exactly what they were dealing with and could act with confidence.


Series Summary

Part Focus Key Takeaway
1 The Myth No single tool does everything. Match the tool to the question.
2 Satellite Good for triage and trends. 10m resolution is coarse for row crops.
3 Drone Excellent detail for blocks. Cannot diagnose without ground truth.
4 Ground The only honest verification. Essential before expensive actions.
5 Playbook Integrate all three: satellite → drone → ground, in that order.

Sources (all 2024-2025): Statistics Canada. (2025). Fruit and vegetable production, 2024. Agriculture and Agri-Food Canada. (2025). Statistical overview of the Canadian fruit industry, 2024. Copernicus. (2024). Sentinel-2 for Agriculture. European Space Agency.


Series complete. Use Part 1 as the hook for social or newsletter distribution, Parts 2-4 as deep-dives for specific grower questions, and Part 5 as the downloadable playbook or reference guide.